Total synthesis of (.+-.)-forskolin

forskolin (mp 199-200 "C). The synthetic material was proven to be identical with an authentic sample of natural forskolin by comparison of the 400 MH...
0 downloads 0 Views 306KB Size
J . Am. Chem. SOC.1988,110, 3672-3673

3672

drolysis, affording the enone 18 as a 64:36 mixture of E to Z isomers in 59% yield. The crucial ring closure of 18 via oxyselenation furnished, after reductive workup, the desired tetrahydropyran-Cone 19 in 78% yield along with 6% yield of its epimer. In stark contrast to the result with a model system,88the stereochemical outcome of this cyclization proved to be independent of the starting olefin geometry (19and its epimer: 79% and 5% from E-18;80% and 6% from 2-18),implying that cyclization proceeded through the chair-preferred transition state involving a stable open carbocation allowing rotation about C12/C-13 bond to direct the methyl group a t an axial position. Transformation of 19 to 20 was quantitatively effected by a well-established Grieco m e t h ~ d . ~ *Sequential + ~ * ~ ~ removal of the p-methoxybenzyl groupZSand acetonide followed by selective acetylationz6 of 7&OH completed the total synthesis of (*)forskolin (mp 199-200 "C). The synthetic material was proven to be identical with an authentic sample of natural forskolin by comparison of the 400 MHz 'H NMR, I3C NMR, IR, MS, and TLC data.z7

Acknowledgment. This research was support4 in part by grants from Japan Research Foundation for Optically Active Compounds and the Ministry of Education, Science, and Culture. We are indebted to N. Matsuura and Y. Yanagiya for their technical assistance and the members of Instrumental Analysis Center of this faculty for spectral measurements. W e are also grateful to Drs. K. Kamiya and Y. Wada of Takeda Chemical Industries, Ltd. for X-ray crystallographic determination. Supplementary Material Available: Spectroscopic data and physical constants for 1-5,7-9,and 11-20 and stereoviews and lists of atomic coordinates, thermal parameters, bond distances, and bond angles for 13 and 16 (21 pages). Ordering information is given on any current masthead page. (24) Grieco, P. A.; Gilman, S.; Nishizawa, M. J . Org. Chem. 1976, 41, 1485. (25) Oikawa, Y.; Yoshioka, T.; Yonemitsu, 0. Terrahedron Len. 1982.23, 885. (26) Bhat, S.V.; Bajwa, B. S.; Dornauer, H.; de Souza, N. J. J . Chem. Soc., Perkin Trans. I 1982, 767. (27) After this paper was submitted, we knew that Ziegler and co-workers developed a synthetic route to forskolin: Ziegler, F. E.; Jaynes, B. H.; Saindane, M. T. J . Am. chem. SOC.1987, 109, 8115.

Total Synthesis of (A)-Forskolin E. J. Corey,* Paul Da Silva Jardine, and John C. Rohloff Department of Chemistry, Harvard University Cambridge, Massachusetts 021 38 Received February 1, 1988

Forskolin (l), a diterpenoid isolated from Coleus forskohlii,' is an activator of adenylate cyclase which has a number of physiological effects (e.g., vaso- and bronchodilating, positive inotropic, and antiglaucoma) and considerable therapeutic potential.z Not surprisingly therefore, many laboratories have embarked on the synthesis of 1. A spate of papers has appeared which describe initial stages of a variety of appro ache^,^ and most (1) (a) Bhat, S. V.; Bajwa, B. S.; Dornauer, H.; de Souza, N. J. Tefrahedron Leri. 1977, 1669-1672. (b) Paulus, E. F. Zeitschrifr fur Krist. 1980, 152, 239-245; 153, 43-49. (c) Bhat, S. V.; Bajwa, B. S.; Dornauer, H. J. Chem. Soc., Perkin Trans. I 1982, 767-771. (2) See: Seamon, K. B. Ann. Rep. Med. Chem. 1984, 19, 293-301. (3) See, for example: (a) Jenkins, P. R.; Menear, K. A.; Barraclough, P.; Nobbs, M. S. J . Chem. Soc., Chem. Commun. 1984, 1423-1424. (b) Nicolaou, K. C.; Li, W. S. Ibid. 1985, 421. (c) Ziegler, F. E.; Jaynes, B. H.; Saindane, M. T. TefrahedronLeu. 1985, 26, 3307-3310. (d) Kulkarni, Y. S.; Snider, B. B. Org. Prep. Prored. Int. 1986, 18, 7-10. (e) Baraldi, P. G.; Barco, A.; Benetti, S.; Pollini, G.P.; Polo, E.; Simoni, D. J. Chem. Soc., Chem. Commun. 1986, 757-758. (0 Hutchinson, J. H.; Pattenden, G.;Myers, P. L. Terrahedron Letr. 1987, 28, 1313-1367. (g) Bold, G.; Chao, S.; Bhide, R.;Wu, S.; Patel, D. V.; Sih, C. J. Ibid. 1987, 28, 1973-1976. (h) Koft, E. R.; Kotnis, A. S.; Broadbent, T. A. Ibid. 1987, 28, 2799-2800.

recently a synthetic pathway has been reported which involves synthesis of a racemic intermediate, partial synthesis of the same intermediate in chiral form from forskolin, and reconversion of the degradation product to f o r ~ k o l i n . ~This paper contains an account of the first total synthesis of (*)-forskolin and a highly enantioselective method for obtaining the first synthetic intermediate 2 in chiral form, so that the approach described herein in principle amounts to a synthesis of the native form of forskolin. The A/B ring system of 1 was constructed simply by allowing hydroxy diene Zsand acetylenic acid 36to react in CHC1, solution (0.44 M) at 23 OC for 30 h to give 4 (72%) as the product of sequential esterification and Diels-Alder reaction. Lactone 4 was transformed into endoperoxide 5 in three steps: (1) replacement of tosyl by methyl (76%) by using 2.7 equiv of MezCuLi and 1.2 equiv of BF3.Etz0 (-35 OC 1 h, to 0 "C 15 min); (2) CY,@ 0,y-double bond isomerization (0.1 equiv of diazabicyclononene (DBN), 23 "C, 45 min); and (3) photoperoxidation of the conjugated diene lactone (Oz, tungsten lamp irradiation, CHC13,0.1% methylene blue; 0 OC, 144 h) to give 5' (95% over two steps). Reduction of 5 (10 equiv of AlHg in 20:l THF-H,O at 23 OC

-

I

4

3

2

COOMe

C6H5C02

5

7

6

)c.'uiO O% 8

9 R = C GCCH2CH20TBMS 10 R = CH=C(OAc)CH,CH ,OTBMS

for 10 min) afforded dihydroxy lactone 6 (97%) which was converted to enone 7 by the following sequence: (1) benzoylation (2 equiv each of benzoic anhydride pyridine, and 4-(dimethylamino)pyridine (DMAP) in CICHzCHzCl at 50 OC for 2 h; 85% yield of 1-monobenzoate); (2) oxidation by pyridinium chlorochromate (9 equiv, ClCH2CHzCl,80-90 OC for 5 h; 60% yield);s (3) lactone reductive cleavage using 13 equiv of AlHg in 20:l THF-HzO at 20 O C for 18 min (85% yield); and (4) esterification with ethereal CH2N2(99%). Lactone acetonide 8 was obtained from 7 in four steps (69% overall): (1) enone and benzoate reduction with lactonization (4.4 equiv of diisobutylaluminum hydride in toluene at -78 OC for 75 min; 80%); (2) stereoselective (4) Ziegler, F. E.; Jaynes, B. H.; Saindane, M. T. J. Am. Chem. Soc. 1987, 109, 8115-8116. ( 5 ) Prepared from a-ionone by the sequence (1) epoxidation by 1.5 equiv of peroxyacetic acid in ethyl acetate (2.9 M) at 23 " C for 3 h (100%); (2) carbonyl reduction using 1 mol equiv of sodium borohydride and 1 equiv of cerium trichloride in methanol at 23 "C for 10 min (100%); (3) ozonolysis in CH30H-CH2C1, followed by treatment with Me2S and subsequent treatment of the aldehyde product with base to afford 2,4,4-trimethyl-3formyl-2-cyclohexen-1-01(90%); and (4) Wittig methylenation in T H F at 0 'C for 1 h (71%). (6) Prepared from ptoluenesulfonylacetylene (Bhattacharya, S . N.; Josiah, B. M.; Walton, D. R. M. Organomer. Chem. Synth. 1970, 1, 145-149) by metalation in T H F at -105 to -95 OC with BuLi (90 min), reaction with excess C 0 2 (-95 "C to 0 "C), acidification and rapid extractive isolation at 0 OC. The acid 3 was used immediately for reaction with 2 since it undergoes rapid (base-catalyzed) decarboxylation. (7) The stereochemistry of 5 was confirmed by the observation of a positive NOE effect between the 8-proton at C(9) and the olefinic protons (at C(6) and C(7)). (8) Dauben, W. G . ; Michno, D.M. J. Org. Chem. 1977, 42, 682-685.

0002-786318811.5 10-3672%01.SO10 0 1988 American Chemical Society

J. Am. Chem. SOC.1988, 110, 3673-3674 7,8-@epoxidation with 2.5 equiv of tert-butyl hydroperoxide and 0.05 equiv of M o ( C O ) ~in C6H6at 68 OC for 1 h;9 (3) elimination of H from C(9) and 0 from C(8) using 4 equiv of KOH in C H 3 0 H at 23 "C for 10 min (86% yield for two steps); and (4) ketalization with excess 2,2-dimethoxypropane-acetonewith tosic acid as catalyst at 23 OC for 90 min (99% yield). The highly reactive lactone carbonyl of 8 was readily ethynylated by slow addition of 2.8 equiv of L i e C C H 2 C H z O T B M S (TBMS = tert-butyldimethylsilyl) to 8 in T H F at 0 OC (go%), and the resulting 1-hydroxy ketone was carbamoylated by reaction with 10 equiv each of dimethylcarbamoyl chloride, 2,6-lutidine, and silver triflate in CHZCl2(0.06 M) at 23 OC (addition of the silver salt to the other two reactants) to give 9 (60%). Ynone 9 was converted to enol acetate 10 (60% overa1l)'O by the following steps: (1) conjugate addition of hydroxyl to C=C (10 equiv of 0.3 M K 2 C 0 3 in 1:l THF-ethylene glycol at 23 OC for 2 h followed by exposure to 1:1 2 N aqueous oxalic acid and acetone at 60 "C for 7 h); (2) resilylation (10 equiv of TBMSC1, 30 equiv of imidazole in D M F at 23 OC for 30 min; 73% overall); and (3) acetylation of the resulting &hydroxy enone by reaction first with thallous ethoxide at 23 OC for 30 min and then acetyl chloride (-78 "C to -45 "C over 1 h; 82%). Irradiation of 10 (GE sunlamp) in the presence of 2% of methylene blue in 0,-saturated CHCl, at 10 OC for 4-5 h resulted in photocyclization to a pyran and subsequent 4 + 2 addition of 'Agoz to form endoperoxide 11 in 55-63% yield. This key step

W

O

W

13 R = CH2CHzOTBMS 14 R = CH= CH,

12

15

h) gave carbonate 15 (>95%). Reaction of 15 with 0.14 M LiOH in 4:2:1 THF-H,O-i-PrOH at 23 OC for 5 min produced (&)desacetyl forskolin (>95%) which upon treatment with excess AqO-pyridine at 0 OC for 4 h gave (A)-forskolin (1) in 90% yield. Synthetic (f)-forskolin thus obtained was identical with an authentic sample of f ~ r s k o l i n by ' ~ 500 M H z 'H N M R , infrared, and high resolution mass spectral comparison as well as by thin layer chromatography by using several different solvent systems. Reduction of the ketone corresponding to 2 by 0.6 equiv of borane in the presence of 10 mol% of the (R)-oxazaborolidine 16 as c a t a l y ~ t ' in ~ ,THF ~ ~ solution proceeded with 9515 enantioselectivity to afford the (S)-enantiomer of 2 (as shown), the form required for enantioselective synthesis of the natural form of forskolin, and this alcohol has been converted to the chiral lactone 4. Thus the synthetic approach reported herein can provide the natural form of forskolin as well as the racemate. A number of the steps of this synthesis are noteworthy or novel including (1) the enantioselective synthesis of 2, (2) the facile one-step synthesis of 4 from 2 and 3 at room temperature, (3) the functional group transformation in the conversion 4 5, 5 7,9 ---+ 11, and 11 12. The stereospecificity of the C-ring annulation 10 12 and the conjugate methylation 12 13 also stand out.16

-

-

+

-

-

Supplementary Material Available: Spectroscopic data for compounds 1-15 and other reaction intermediates mentioned herein (4 pages). Ordering information is given on any current masthead page. (13) We thank Drs. R. H. Rupp, W. Bartmann, and J. Knolle of the Hoechst Co. for a generous supply of plant-derived forskolin. (14) Corey, E. J.; Bakshi, R.; Shibata, S.; Chen, C.-P.; Singh, V. K. J. Am. Chem. SOC.1987, 109, 7925-7926. (15) We are indebted to Dr. Tetsuya Mohri of these laboratories for carrying out the enantioselective synthesis of 2 which will be reported separately. (16) We are grateful to the National Institutes of Health for support of this research and to Dr. Keith Kyler for experimental contributions to an early stage of our program.

O

O1;

O% 11

3673

16

to form the C ring of the forskolin system was completely stereoselective." Enone 12 was obtained from endoperoxide 11 by the following sequence: ( 1) b-elimination-hydroperoxidereduction using sodium ethoxide (0.05 M, 2.2 equiv)-tributylphosphine (10 equiv) in ethanol at 0 OC for 2.5 h (80%) and (2) cyclic carbonate formation by reaction with 10:1 acetic acid-acetic anhydride at 100-105 O C (sealed tube) for 23 h. @-Face stereospecificconjugate addition of methyl to enone 12 was effected by reaction with excess MeCuPBu, and BF3-Et20(each 0.2 M ) in ether at -78 "C for 4 h and -50 OC for 15 min to provide keto carbonate 13 in 85% yield. Conversion of 13 to vinyl ketone 14 was carried out in >90% yield by (1) desilylation using 2% HF in 50: 1 acetonitrile-water at 0 OC for 15 min, (2) reaction with o-nitrophenylselenocyanide12 and tri-n-butylphosphine (each 0.02 M) in THF at 0 "C for 2 h, and (3) treatment with 10 equiv of 30% aqueous hydrogen peroxide in THF (0.16 M) at 23 OC for 4 h. Deketalization of 14 (2:l acetic acid-water, 10 equiv of semicarbazide, 70 OC, 4 (9) Sharpless, K. B.; Michaelson, R. C. J . Am. Chem. SOC.1973, 95, 6 1 36-6 1 37. (10) The 'HN M R spectrum of 10 indicates rapid interconversion of the two position isomeric 8-acetoxyenones (acetyl migration) at 23 OC which becomes slow on the N M R time scale at 227 K. (1 1) For related photocyclizations, see: (a) Buchi, G.; Yang, N. C. Helu. Chim. Acta 1955, 38, 1338-1341. (b) Cerfontain, H.; van Noort, P. C . M.; Geenevasen, J. A. J. J . Chem. SOC.,Perkin Trans. I I 1980, 1057-1062. (c) Barker, A. J.; Begley, M. J.; Mellor, M.; Otieno, D. A.; Pattenden, G. J . Chem. SOC.,Perkin Trans. I1983, 1893-1900. (12) (a) Sharpless, K. B.; Young, M. W. J . Org. Chem. 1975,40,947-949. (b) Grieco, P. A.; Gilman, S.; Nishizawa, M. J . Org. Chem. 1976, 41, 1485-1486.

0002-7863/88/1510-3673$01.50/0

Hexagonal Lattice Hosts for Urea. A New Series of Designed Heterocyclic Receptors Thomas W. Bell* and Jia Liu Department of Chemistry State University of New York Stony Brook, New York 1 1794-3400 Received November 27. 1987 Beginning with crown ethers,' the field of host-guest,, or sup r a m o l e c ~ l a r ,chemistry ~ focused initially on complexation of cation^.^ Although hydrogen bonds between neutral molecules are generally weaker than charge/dipole attraction and polar hydrogen bonds,5 several recent reports indicate that networks of hydrogen bonds may be used to form neutral complexes that ( 1 ) Pedersen, C. J. J . Am. Chem. SOC.1967, 89, 2495-2496.

(2) Cram, D . J. Angew. Chem., Int. Ed. Engl. 1986, 25, 1039-1057, and references therein. (3) Lehn, J. M. Science (Washington, D.C.)1985, 227, 849-856, and references therein. (4) For additional reviews, see: (a) Sutherland, I. 0 . Chem. SOC.Rev. 1986, 15, 63-91. (b) Colquhoun, H. M.; Stoddart, J. F.; Williams, D. J. Angew. Chem., Int. Ed. Engl. 1986, 25, 487-507. (c) Synthesis of Macrocycles. Progress in Macrocyclic Chemistry; Izatt, R. M.; Christensen, J. J. Eds.; Wiley: New York, 1987; Vol. 3. (5) (a) Saenger, W. Principles of Nucleic Acid Structure; Springer-Verlag: New York, 1984; Chapter 6. (b) Meot-Ner (Mautner), M. In Molecular Structure and Energetics, Vol. 4: Biophysical Aspects; Liebman, J . F., Greenberg, A,, Eds.; VCH: New York, 1987; p 71.

0 1988 American Chemical Society